A series of studies are proposed to deal with the types of codes used in both reading and scene perception. The studies deal with how visual, orthographic, phonological, and semantic codes are used in foveal and parafoveal vision. in most of the experiments, subjects' eye movements will be monitored in normal reading (or reading related tasks) or during scene perception. The primary technique used in the experiments involves making display changes in either the text or scene contingent upon the position of the eyes. Eye movement data (fixation time, saccade length, and number of fixations) will be examined along with other measures (such as reading rate, scene viewing time, or naming time) to determine (1) how different codes are used in identifying word and objects, (2) how different codes are used in integrating information across saccades, and (3) the relationship between attention and eye movements. The primary purpose of the research is to develop more sophisticated models of the reading process and of scene perception. The experiments will also help to determine how the perceptual processes involved in reading and scene perception are similar or different. Finally, some of the experiments examine beginning and dyslexic readers and the research should further help us to understand reading in both groups.